Annual Forum 2004 – Molecular Targets in Pediatric Malignancies

Gary Gilliland, Ph.D., MD, Howard Hughes Medical Institute, Boston, MA
Charles Sawyers, MD, University of California, Los Angeles, CA

Dr. Sara A. Courtneidge, Van Andel Research Institute, Grand Rapids, MI
David E. Fisher, MD, Ph.D., Dana Farber Cancer Institute, Boston, MA
Todd R. Golub, MD, Dana Farber Cancer Institute, Boston, MA
Andrei V. Gudkov, Ph.D., D.Sci., The Cleveland Clinic Foundation, Cleveland, OH
Dr. Patrick Harran, University of Texas - Southwestern Medical Center, Dallas, TX
William Kaelin, MD, Dana-Farber Cancer Institute, Boston, MA
John M. Maris, MD, Children’s Hospital of Philadelphia, Philadelphia, PA
William R. Sellers, MD, Dana Farber Cancer Inst, Boston, MA
Kevin Shannon, MD, University of California, San Francisco, CA
Gregory L. Verdine, Harvard University, Cambridge, MA
Marc Vidal, Dana Farber Cancer Institute, Boston, MA

Layout and Goals

I. TK and STK as targets in Therapy in Cancer
II. Novel Approaches to Drug Development
III. From Kinases to Oxygen Sensing: Novel Targets for Cancer Therapy
IV. Novel Approaches to Pediatric Malignancies

This topic pertains to the new quest of utilizing molecular and genetic information about cancer to specifically design treatments which target functionally important molecular lesions. In the past several decades enormous quantities of information have accumulated regarding the precise mutations which either activate or disrupt genes in many human cancers. In some cases this information has permitted new classification of cancers, and the molecular abnormalities have become central to the diagnosis of specific tumors. While large strides have been made in the identification of gene abnormalities in cancer, therapeutic advances have not been as forthcoming. This stems largely from the fact that traditional cancer therapies have been employed prior to a complete understanding of the mechanisms through which they work or the cellular targets which they attack. The "Targeted Therapy" concept stems from the goal of developing cancer treatments which specifically focus on disrupting molecules within cancer cells which are known to be central to the malignant behavior of the cell. Such targets have included activated oncogenes such as Abl, a factor whose inhibition by the drug Gleevec has produced dramatic remissions in certain forms of leukemia. The discussion at this Forum will focus both on the identification of targets as well as in strategies to design drugs capable of inhibiting them in order to convert the information learned about cancer into therapeutic advances.

Outcome Report
Several themes were developed during the course of the symposium. These included the concept of “oncogene addiction” in which a cancer cell becomes “addicted” to its mutant oncogene, rendering it susceptible to induction of cell death by oncogene inhibitors. The role of the microenvironment in supporting tumor cell growth, and as a target for therapy was discussed. In addition to targeting of kinases in cancer, such as BCR-ABL, the potential for targeting the apoptotic machinery of cancer cells was presented by several investigators. Novel approaches to drug development were reviewed by experts in organic and medicinal chemistry, as were novel approaches to understanding pathogenesis of cancer and development of new targets for treatment based on current genomic technologies and the use of network analysis. It was noted that our data sets for understanding cancer biology at the genomic, biochemical, cellular and organismal level are incomplete, but that there is potential in the coming years to develop such data sets using sophisticated technological approaches.

The heterogeneity of tumors was discussed in the context of drug development, including the concept of “cancer stem cells” as rare but critical targets among the larger tumor burden in cancer patients. Finally, sociopolitical issues that related to cancer therapy were overviewed, including a discussion of the role that academic institutions are best able to play. These include mandates to discover new targets and to break paradigms in drug development currently held in industry. The development of selective small molecule kinase inhibitors such as imatinib in collaboration between industry and academics serves as a proof-of-principle for this latter approach, and promising approaches to develop small molecules that interfere with protein-protein interaction interfaces may serve as a next frontier. In addition, strategies for ensuring development of pediatric as well as adult drug trials were considered, as well as the most efficient approaches for the academic-industry interface in cancer drug development.

Dr. David Fisher provided an update on the role of B-RAF in cancer, noting that the V600E allele has been identified in 70% of cases of melanoma, 10% of colon cancer, and 33% of papillary thyroid cancer. The remaining cases of melanoma have activating mutation in N-RAS, and thus these signal transduction pathways may serve as useful targets for therapeutic intervention. Dr. Fisher also presented data on development of a zebrafish model of melanoma induced by expression of B-RAF as a reagent to better understand modifiers of phenotype.

The role of SRC in a spectrum of cancers was discussed by Dr. Sara Courtneidge, who noted that SRC is often overexpressed, but only rarely mutated in cancer. There are several good SRC inhibitors currently available, begging the question of the best clinical context for application/clinical trials. She also discussed the role of the microenvironment in tumorigenesis, including the characterization of the locally invasive properties of cancer cells mediated by podosomes. Proteins that are responsible for podosome formation, including a novel protein named “FISH” for five SH3 domains were discussed.

Dr. Nabeel Bardeesy provided an overview of the molecular pathogenesis of the Peutz-Jeughers cancer syndrome associated with pancreatic adenocarcinoma and colonic polyposis that is curiously caused by loss, rather than gain of function, in the serine threonine kinase LKB1. The notion that such mutations confer an inability of cells to manage energy stress response was discussed, and the paradox of how such a mutation confers a proliferative/survival advantage to cancer cells. Imatinib resistance has emerged as an important problem in treatment of BCR-ABL positive CML, and as first described by Dr. Charles Sawyers, is most often attributable to point mutations in the ABL kinase itself. Dr. Sawyers presented exciting new data demonstrating preclinical and clinical efficacy of a novel ABL and SRC inhibitor, BMS354825 that overcomes resistance to imatinib for all BCR-ABL mutations except the T315I. This compound was reported to show promising results in clinical trials, and provides a paradigm for developing drugs that can overcome resistance to small molecule kinase inhibitors as single agent.

The challenges of developing drugs for targets that are discovered in cancer was reviewed by Dr. Greg Verdine, who annotated the “Lipinsky rules” for drug development that mandate that compounds be below a threshold molecular weight, have certain solubility characteristics, etc. Dr. Verdine suggested that such constraints needed to be re-evaluated, and noted a gap in what he referred to as “macrosynthetic land” in drug development. For example, alpha-helical peptide structures present major barriers in drug development due to size and predilection for proteolytic cleavage, but Dr. Verdine and his collaborators have developed strategies using “hydrocarbon staples” to stabilize such structures. These may have value in targeting the apoptotic machinery of cancer cells. Such advances provide a superb example of how academic investigators can challenge dogma in drug development, and potentially move the field toward developing novel agents. In a similar vein, Dr. Patrick Harran described the development of a dimeric small molecule with the ability to impair the normal function of the IAP family of proteins through disruption of the SMAC/IAP protein-protein interface. This discovery provides proof-of-principle that small molecules can indeed disrupt protein-protein interactions mediated by large surface interfaces. Such “SMAC” mimetic compounds may also be of value in targeting apoptotic machinery in cancer cells.

Another strategy for targeting the apoptotic machinery was presented by Dr. Tony Letai. Dr. Letai has developed a peptide “BAD BH3 sensitizer” that induces apoptosis through cytochrome C release in cancer cells but not normal cells. Such compounds have the potential to induce death in cancer cells by targeting any of a number of proapoptotic BH3 containing proteins like BAD.

The discussion then moved from chemistry and a focus on targeting the apoptotic machinery in cancer cells to the use of genome wide strategies for characterizing cancer, for target gene discovery, and drug development. Dr. Marc Vidal presented sophisticated new technology for developing complex networks as probes for differences between normal and cancer cells, using frameworks referred to as “interactomes, phenomes, transcriptomes and localazomes.” He likened this approach as similar to the development of supercomputing 50 years ago, as a tool that is computationally based, but should eventually allow us to understand the very complex networks at a cellular level that will enable distinctions between cancer cells and normal cells. Dr. Bill Sellers discussed the heterogeneity of cancer, and the need to break down cancers that may be similar phenotypically into subgroups that are homogeneous genetically. As an example he noted that IGF-1 is a marker in prostate cancer that is associated with a lethal outcome, and thus serves as a tool for targeting that subset of prostate cancer with novel therapeutics, as well as allowing for stratification of response to conventional therapies. He noted that with modern genomic technologies, it should be possible to generate a complete genetic subclassification of cancer that can interface with therapeutic efficacy using response to drugs as a probe for additional subclassification.

Examples of the remarkable power of cancer genomics were provided by Dr. Todd Golub, who discussed the use of chemical genetic screens for compounds that can induce differentiation in acute leukemia; in determining that the metastatic potential of a tumor resides in all cells in the primary tumor; and in assessing cell autonomous and cell non-autonomous contributions to cancer pathogenesis. He noted a lack of complete data sets in understanding cancer-drug interactions, and presented a strategy for testing panels of molecularly targeted therapies in a broad spectrum of cancer cell lines as an approach to understand pathophysiology of cancer, as well as for developing novel therapeutic approaches.

Dr. Kim Rathmell presented data on the interesting relationship between oxygen sensing and cancer derived from study of the Von-Hippel Lindau (VHL) gene that is mutant in VHL syndrome that is associated with hemangioblastoma, pheochromocytoma and renal cell carcinomas. Dr. Rathmell characterized the effect various VHL mutations observed in humans with expression of the hypoxia inducible factors HIF1 and HIF2, and suggested that molecularly targeted therapies for this pathway, directed to e.g. HIF, VEGF or COX2 could be tested using ES cells expressing various VHL mutants.

The concept of cancer stem cells, with a focus on leukemia stem cells, was presented by Dr. Gary Gilliland. He noted only a small fraction of leukemia cells (~1 in 100,000) has the capacity for limitless self-renewal, and that it is thought to be these cells that are required for continued growth and propagation of leukemia and other tumor types such as breast cancer and CNS tumors. He noted that tumor relapse after initial response to therapy may be due to failure to adequately target these rare populations of cells with unlimited self-renewal capacity. He also suggested that targeting self-renewal pathways in this critical population of cells might be of therapeutic value.

Dr. Bill Kaelin reviewed problems with current approaches to drug development - including a discussion of the value of murine disease models in preclinical platforms. Dr. Kaelin noted that although tumors may be quite complex phenotypically and genetically, and harbor a broad spectrum of mutations, they may nonetheless be exquisitely sensitive to inhibition of a single mutant allele. As an example, he noted the remarkable responses of genetically complex gastrointestinal stromal cell (GIST) tumors with activating mutations in the tyrosine kinase KIT to imatinib. Dr. Kaelin also noted the importance of diligent and comprehensive searches for good drug targets in academic contexts. He emphasized the importance of thoughtful clinical trial design for newer molecularly targeted therapies, especially as regards rapid progress toward FDA approval.

Dr. John Maris focused attention on pediatric cancers, in particular on recent developments in our understanding of the molecular pathogenesis of neuroblastoma. He noted that a new heritable predisposition gene/locus has been identified that may shed new light on pathogenesis of neuroblastoma, and discussed recent data from animal models of disease using a novel drug METAP2. He noted that an important problem in pediatric cancer was not the lack of patients for treatment on clinical trials, but rather the lack of good targets identified thus far. Dr. Maris also noted potential for use of TRK kinase inhibitors for treatment of neuroblastoma that express these proteins, including the TRK inhibitor CEP-701 that is currently being tested in clinical trials in AML based on its ability to also inhibit FLT3.

Dr. Andrei Gudkov discussed p53 induced arrest and irradiation, and the role of p53 as a negative regulator of mitotic catastrophe. In summary, a broad spectrum of topics related to cancer genetics and cancer therapeutics were overviewed. Hope and optimism was expressed for the opportunities to develop novel drugs to target cancers based on state-of-the art screening and cancer genomics strategies, both in pediatric and adult tumors. These was an overall consensus that proof-of-principle had been demonstrated for development of molecularly targeted therapies, and that is was likely that major progress would be made in cancer therapeutics in the coming years.